India-born scientist"s Robo Brain is a very fast online learner

August 25, 2014

Robo BrainMumbai, Aug 25: In July, scientists from Cornell University led by Ashutosh Saxena said they have developed Robo Brain—a large computational system that learns from publicly available Internet resources. The system, according to a 25 August statement by Cornell, is downloading and processing about 1 billion images, 120,000 YouTube videos and 100 million how-to documents and appliance manuals.

Information from the system, which Saxena had described at the 2014 Robotics: Science and Systems Conference in Berkeley, is being translated and stored in a robot-friendly format that robots will be able to draw on when needed.

The India-born, Indian Institute of Technology-Kanpur graduate, has now launched a website for the project at robobrain.me, which will display things the brain has learnt, and visitors will be able to make additions and corrections. Like a human learner, Robo Brain will have teachers, thanks to crowdsourcing. “Our laptops and cellphones have access to all the information we want.

If a robot encounters a situation it hasn"t seen before it can query Robo Brain in the cloud,” Saxena, assistant professor, Microsoft Faculty Fellow, and Sloan Fellow, at Cornell University, said in a statement.

Saxena and his colleagues at Cornell, Stanford and Brown universities and the University of California, Berkeley, say Robo Brain will process images to pick out the objects in them, and by connecting images and video with text, it will learn to recognize objects and how they are used, along with human language and behaviour.

His team includes Ashesh Jain, a third-year PhD computer science student at Cornell. Robo Brain employs what computer scientists call structured deep learning, where information is stored in many levels of abstraction.

Deep learning is a set of algorithms, or instruction steps for calculations, in machine learning. For instance, an easy chair is a member of a class of chairs, and going up another level, chairs are furniture.

Robo Brain knows that chairs are something you can sit on, but that a human can also sit on a stool, a bench or the lawn, the statement said.

A robot"s computer brain stores what it has learnt in a form that mathematicians call a Markov model, which can be represented graphically as a set of points connected by lines—called nodes and edges.

The nodes could represent objects, actions or parts of an image, and each one is assigned a probability—how much you can vary it and still be correct.

In searching for knowledge, a robot"s brain makes its own chain and looks for one in the knowledge base that matches within those limits.

“The Robo Brain will look like a gigantic, branching graph with abilities for multi-dimensional queries,” said Aditya Jami, a visiting researcher art Cornell, who designed the large database for the brain. Jami is also co-founder and chief technology officer at Predict Effect, Zoodig Inc. The basic skills of perception, planning and language understanding are critical for robots to perform tasks in the human environments. Robots need to perceive with sensors, and plan accordingly.

If a person wants to talk to a robot, for instance, the robot has to listen, get the context and knowledge of the environment, and plan its motion to execute the task accordingly.

For example, an industrial robot needs to detect objects to be manipulated, plan its motions and communicate with the human operator. A self-driving robot needs to detect objects on the road, plan where to drive and also communicate with the passenger.

Scientists at the lab at Cornell do not manually programme the robots. Instead, they take a machine learning approach by using variety of data and learning methods to train our robots.

“Our robots learn from watching (3D) images on the Internet, from observing people via cameras, from observing users playing video games, and from humans giving feedback to the robot,” the Cornell website reads.

There have been similar attempts to make computers understand context and learn from the Internet.

For instance, since January 2010, scientists at the Carnegie Mellon University (CMU) have been working to build a never-ending machine learning system that acquires the ability to extract structured information from unstructured Web pages.

If successful, the scientists say it will result in a knowledge base (or relational database) of structured information that mirrors the content of the Web. They call this system the never-ending language learner, or NELL.

NELL first attempts to read, or extract facts from text found in hundreds of millions of web pages (plays instrument). Second, it attempts to improve its reading competence, so that it can extract more facts from the Web, more accurately, the following day. So far, NELL has accumulated over 50 million candidate beliefs by reading the Web, and it is considering these at different levels of confidence, according to information on the CMU website.

“NELL has high confidence in 2,348,535 of these beliefs—these are displayed on this website. It is not perfect, but NELL is learning,” the website reads.

We also have IBM, or International Business Machines" Watson that beat Jeopardy players in 2011, and now has joined hands with the United Services Automobile Association (USAA) to help members of the military prepare for civilian life.

In January 2014, IBM said it will spend $1 billion to launch the Watson Group, including a $100 million venture fund to support start-ups and businesses that are building Watson-powered apps using the “Watson Developers Cloud”.

More than 2,500 developers and start-ups have reached out to the IBM Watson Group since the Watson Developers Cloud was launched in November 2013, according to a 22 August blog in the Harvard Business Review.

Comments

Add new comment

  • Coastaldigest.com reserves the right to delete or block any comments.
  • Coastaldigset.com is not responsible for its readers’ comments.
  • Comments that are abusive, incendiary or irrelevant are strictly prohibited.
  • Please use a genuine email ID and provide your name to avoid reject.
News Network
July 9,2020

U.S. electric vehicle maker Tesla Inc is "very close" to achieving level 5 autonomous driving technology, Chief Executive Elon Musk said on Thursday, referring to the capability to navigate roads without any driver input.

"I'm extremely confident that level 5 or essentially complete autonomy will happen and I think will happen very quickly," Musk said in remarks made via a video message at the opening of Shanghai's annual World Artificial Intelligence Conference (WAIC).

"I remain confident that we will have the basic functionality for level 5 autonomy complete this year."

Automakers and tech companies including Alphabet Inc Waymo and Uber Technologies are investing billions in the autonomous driving industry.

However industry insiders have said it would take time for the technology to get ready and public to trust autonomous vehicles fully.

The California-based automaker currently builds cars with an Autopilot driver-assistance system.

Tesla is also developing new heat-projection or cooling systems to enable more advanced computers in cars, Musk said.

Industry data showed Tesla sold nearly 15,000 China-made Model 3 sedans last month.

Tesla has become the highest-valued automaker as its shares surged to record highs and its market capitalisation overtook that of former front-runner Toyota Motors Corp.

Comments

Add new comment

  • Coastaldigest.com reserves the right to delete or block any comments.
  • Coastaldigset.com is not responsible for its readers’ comments.
  • Comments that are abusive, incendiary or irrelevant are strictly prohibited.
  • Please use a genuine email ID and provide your name to avoid reject.
News Network
April 17,2020

Paris, Apr 17: Even as virologists zero in on the virus that causes COVID-19, a very basic question remains unanswered: do those who recover from the disease have immunity?

There is no clear answer to this question, experts say, even if many have assumed that contracting the potentially deadly disease confers immunity, at least for a while.

"Being immunised means that you have developed an immune response against a virus such that you can repulse it," explained Eric Vivier, a professor of immunology in the public hospital system in Marseilles.

"Our immune systems remember, which normally prevents you from being infected by the same virus later on."

For some viral diseases such a measles, overcoming the sickness confers immunity for life.

But for RNA-based viruses such as Sars-Cov-2 -- the scientific name for the bug that causes the COVID-19 disease -- it takes about three weeks to build up a sufficient quantity of antibodies, and even then they may provide protection for only a few months, Vivier told AFP.

At least that is the theory. In reality, the new coronavirus has thrown up one surprise after another, to the point where virologists and epidemiologists are sure of very little.

"We do not have the answers to that -- it's an unknown," Michael Ryan, executive director of the World Health Organization's Emergencies Programme said in a press conference this week when asked how long a recovered COVID-19 patient would have immunity.

"We would expect that to be a reasonable period of protection, but it is very difficult to say with a new virus -- we can only extrapolate from other coronaviruses, and even that data is quite limited."

For SARS, which killed about 800 people across the world in 2002 and 2003, recovered patients remained protected "for about three years, on average," Francois Balloux director of the Genetics Institute at University College London, said.

"One can certainly get reinfected, but after how much time? We'll only know retroactively."

A recent study from China that has not gone through peer review reported on rhesus monkeys that recovered from Sars-Cov-2 and did not get reinfected when exposed once again to the virus.

"But that doesn't really reveal anything," said Pasteur Institute researcher Frederic Tangy, noting that the experiment unfolded over only a month.

Indeed,several cases from South Korea -- one of the first countries hit by the new coronavirus -- found that patients who recovered from COVID-19 later tested positive for the virus.

But there are several ways to explain that outcome, scientists cautioned.

While it is not impossible that these individuals became infected a second time, there is little evidence this is what happened.

More likely, said Balloux, is that the virus never completely disappeared in the first place and remains -- dormant and asymptomatic -- as a "chronic infection", like herpes.

As tests for live virus and antibodies have not yet been perfected, it is also possible that these patients at some point tested "false negative" when in fact they had not rid themselves of the pathogen.

"That suggests that people remain infected for a long time -- several weeks," Balloux added. "That is not ideal."

Another pre-publication study that looked at 175 recovered patients in Shanghai showed different concentrations of protective antibodies 10 to 15 days after the onset of symptoms.

"But whether that antibody response actually means immunity is a separate question," commented Maria Van Kerhove, Technical Lead of the WHO Emergencies Programme.

"That's something we really need to better understand -- what does that antibody response look like in terms of immunity."

Indeed, a host of questions remain.

"We are at the stage of asking whether someone who has overcome COVID-19 is really that protected," said Jean-Francois Delfraissy, president of France's official science advisory board.

For Tangy, an even grimmer reality cannot be excluded.

"It is possible that the antibodies that someone develops against the virus could actually increase the risk of the disease becoming worse," he said, noting that the most serious symptoms come later, after the patient had formed antibodies.

For the moment, it is also unclear whose antibodies are more potent in beating back the disease: someone who nearly died, or someone with only light symptoms or even no symptoms at all. And does age make a difference?

Faced with all these uncertainties, some experts have doubts about the wisdom of persuing a "herd immunity" strategy such that the virus -- unable to find new victims -- peters out by itself when a majority of the population is immune.

"The only real solution for now is a vaccine," Archie Clements, a professor at Curtin University in Perth Australia, told AFP.

At the same time, laboratories are developing a slew of antibody tests to see what proportion of the population in different countries and regions have been contaminated.

Such an approach has been favoured in Britain and Finland, while in Germany some experts have floated the idea of an "immunity passport" that would allow people to go back to work.

"It's too premature at this point," said Saad Omer, a professor of infectious diseases at the Yale School of Medicine.

"We should be able to get clearer data very quickly -- in a couple of months -- when there will be reliable antibody tests with sensitivity and specificity."

One concern is "false positives" caused by the tests detecting antibodies unrelated to COVID-19.

The idea of immunity passports or certificates also raises ethical questions, researchers say.

"People who absolutely need to work -- to feed their families, for example -- could try to get infected," Balloux.

Comments

Add new comment

  • Coastaldigest.com reserves the right to delete or block any comments.
  • Coastaldigset.com is not responsible for its readers’ comments.
  • Comments that are abusive, incendiary or irrelevant are strictly prohibited.
  • Please use a genuine email ID and provide your name to avoid reject.
Agencies
January 20,2020

Washington D.C., Jan 20: An American bride asked for money from her invitees so that they can be on the 'exclusive guest list'.

Weddings can be surely expensive. But is it feasible for one to charge the guests to make up for the expenses?

According to Fox News, that is exactly what happened in a recent American wedding. A 19-year-old shared on Reddit that her cousin was getting married on Sunday and announced that she would charge 50 dollars to those who wanted to attend her wedding.

"She said that they can Venmo her money so there won't be no [sic] problems and everyone who paid will be added onto the 'exclusive guest list' which basically means you won't have to wait in line while other guests pay," wrote the user named DaintySheep.

While she refused to pay for entry into her cousin's wedding the bride-to-be contacted the elders in the family which ended up in an embarrassing situation.

"She wanted to get the money she spent on her special day back. I told her I wouldn't be able to come because this was outrageous and that I wish her well on her special day. She contacted my aunt and my aunt called me cheap and rude. My parents offered to pay for my entry, but I refused," continued the disheartened girl.

While in almost every nook and cranny of the world gifting the bride-groom with money is a tradition, asking for money from friends and family to replenish the money spent on a wedding is can be said to be a rare scenario.

Comments

Add new comment

  • Coastaldigest.com reserves the right to delete or block any comments.
  • Coastaldigset.com is not responsible for its readers’ comments.
  • Comments that are abusive, incendiary or irrelevant are strictly prohibited.
  • Please use a genuine email ID and provide your name to avoid reject.